PROJECT SUMMARY The proposed phased innovation grant will test personalized cognitive bias modification as a novel intervention to reduce interpretation bias and, subsequently, anxiety in clinically anxious youth. Anxiety is the most common mental health problem across development, affecting one in four children and adolescents. In response to data that evidence-based interventions are ineffective for a substantial proportion (~50%) of youth, experts have called for development of novel treatments that directly target underlying mechanisms. One such promising mechanism is interpretation bias, the inaccurate attribution of threat when faced with ambiguity. We have previously demonstrated that interpretation bias is a ubiquitous marker of anxiety in children and adolescents, and therefore may be an ideal treatment target. Cognitive bias modification for interpretation bias (CBM-I) is a computerized intervention that attempts to reduce anxiety symptoms by directly reducing interpretation bias. In CBM-I, participants are provided with real-time corrective feedback when they select threat-relevant interpretations of ambiguity, with the expectation that they will learn to automatically make more neutral (rather than threat) interpretations over time. CBM-I has demonstrated preliminary efficacy in adults for reducing anxiety symptoms. Yet extant data on CBM-I in clinically anxious youth are sparse, with little examination of interpretation bias as a malleable target and mechanism of action. To date, several critical questions remain to be answered: 1) is the targeted mechanism (interpretation bias) modified by CBM-I in youth?; 2) what dose reflects meaningful interpretation bias change?; and 3) does anxiety reduce because of reductions in interpretation bias? In response to RFA-MH-18-704, we use an experimental therapeutics approach to test interpretation bias as a putative intervention target with CBM-I. This proposal includes a two-phased study of personalized CBM-I in youth ages 10 to 17 who meet diagnostic criteria for anxiety disorder (generalized, social, separation). In the R61 Phase (N=36), a randomized controlled trial (RCT) examines whether and at what dose CBM-I personalized to youths? individual anxiety symptoms reduces interpretation bias compared to a matched computerized interpretation control condition (ICC). In the R33 phase, an RCT (N=66) validates target engagement and conducts a mechanism test (i.e., whether anxiety reduction is due to interpretation bias reduction) by comparing personalized CBM-I to cognitive restructuring, a clinically relevant psychosocial intervention that also targets anxious cognition. Strategic design decisions based on the extant literature in both adults and youth will maximize the likelihood of target engagement. If successful, this experimental intervention could represent a significant step forward in developing personalized treatments that directly modify processes underlying youth anxiety, with potential for clinical translation into real-world clinical and community settings.